skills-cli
Skills CLI
Use bunx skills commands to manage skills.
Workflow
- Inspect current state with
bunx skills listfor project scope andbunx skills list -gfor global scope. - Use
bunx skills find <query>orbunx skills add <source> --listto discover options. - Install with explicit scope and agent targeting:
- For this repository's published skills, use
https://github.com/sjunepark/custom-skills/tree/main/skillsrather than the repo root or current working tree so installs can update across machines without exposing repo-local.agents/and.claude/skills. - For this repo's normal machine-global setup, use
--copy -g -a claude-code -a piso the skills live directly in~/.claude/skillsand~/.pi/agent/skillsandskills list -greportsAgents: Claude Code, Pi. - Shared
~/.agents/skillsinstalls are for intentional universal multi-harness sharing; they can makeskills list -greport many agents for one skill. - Project setup: omit
-g.
- For this repository's published skills, use
- Verify install with the matching scope command and test a prompt that should trigger the skill.
- For generic updates, run
bunx skills checkthenbunx skills update. For this repo's published skills, prefer reinstall-from-remote when converting from shared~/.agents/skillsinstalls to Claude Code + Pi only. - For cleanup, use
bunx skills remove ...with matching scope/agent flags.
Key commands
bunx skills add <package-or-url>bunx skills add <source> --skill <name> -a <agent> [-g] -ybunx skills add <source> --allbunx skills add <source> --skill '*' --copy -g -a claude-code -a pi -ybunx skills list [-g] [-a <agent>]bunx skills find [query]bunx skills remove [skill...] [-g] [-a <agent>]bunx skills checkbunx skills updatebunx skills init [name]bunx skills experimental_installbunx skills experimental_sync
Guardrails
- Prefer explicit
--agentand--skillflags for reproducibility. - Use
--yesonly for non-interactive runs. - Default to symlink mode unless the user requests
--copyor the goal is this repo's normal machine-global Claude Code + Pi setup. bunx skills listwithout-greports project-visible skills for the current directory; do not use it to answer what is globally installed.- When installing this repository's skills for regular machine use, do not install from
.or./skills, and do not use the repo root URL; usehttps://github.com/sjunepark/custom-skills/tree/main/skills. - If the user wants
skills list -gto show onlyAgents: Claude Code, Pifor this repo's published skills, do not leave those installs in~/.agents/skills; install them with--skill '*' --copy -g -a claude-code -a pi -y. - Do not use
--allfor that setup. In the currentskillsCLI,--allexpands to--skill '*' --agent '*' -y, which overrides the Claude Code + Pi restriction and recreates shared~/.agents/skillsinstalls. - Treat installed skills as executable instructions; avoid untrusted sources.
- If managing dotfiles with chezmoi, avoid
chezmoi addon live skills directories.
References
Read references/cli.md for source formats, scopes, and practical command recipes.
More from sjunepark/custom-skills
summarize
Use the steipete/summarize CLI to summarize URLs, local files, stdin, YouTube links, podcasts, and media with LLM models. Trigger when users ask to install or run summarize, configure model/provider API keys, tune output flags (length/language/json/extract/slides), set defaults in ~/.summarize/config.json, or troubleshoot summarize CLI errors.
41post-implementation-review
Manually review already-implemented code for design flaws, abstraction issues, structural problems, or refactors that only became clear in real code. Use only when the user explicitly asks for a post-implementation review, explicitly asks whether recent implementation work revealed design or structure problems, or explicitly wants refactor recommendations after the code exists. Do not auto-trigger for ordinary implementation, debugging, explanation, or generic code review requests. Prefer embedded snippets with file-path comments over editor-oriented file and line references. Prioritize real design, ownership, abstraction, and organization weaknesses over minor local polish, call out real tradeoffs, and separate straightforward recommendations from items worth discussion.
29architecture-md-writer
Create, update, review, and split ARCHITECTURE.md files that explain a codebase's shape, major components, runtime flow, code map, and important invariants. Use when a repository lacks architecture docs, an existing ARCHITECTURE.md is stale or too detailed, a subsystem needs its own nested ARCHITECTURE.md, or a root architecture doc should link to deeper module architecture docs.
27agents-md-writer
Create, edit, review, and improve AGENTS.md files for repositories used by agentic coding tools with concise, actionable instructions and correct precedence behavior. Use whenever AGENTS.md content is being changed, including updating existing guidance, drafting a new AGENTS.md, migrating legacy instruction files, defining nested overrides in monorepos, or debugging why tools load the wrong guidance.
26source-investigator
Investigate external libraries, frameworks, and unfamiliar repositories by cloning the exact repo into a project-local temp workspace, ignoring that workspace in git, and delegating code reading to focused subagents so the main thread stays clean. Use whenever docs are incomplete, version-specific behavior matters, you need to learn how a codebase works, or exploring lots of source inline would pollute the main context.
24briefing
Manually brief the user on the current task or session so they can make the next decision without reloading the whole codebase. Use only when the user explicitly asks for a briefing, catch-up, current state, relevant architecture, or implementation context. Prefer this skill for task-scoped context recovery when the conversation is long, the code has changed, or the user wants the relevant details without a long explanation.
22